Concerns About Crime for Self and Others

Abstract

The purpose of this study was to examine how individual- and contextual-level factors affect both personal concerns about crime and concerns about crime for one’s spouse or significant other, family, and friends. We were also interested in whether and how these factors differed by gender. We combined individual-level data from a telephone survey of U.S. adults (N = 1,207) with contextual-level data from the U.S. Census, the American Community Survey, and the Uniform Crime Reports. We found that respondents were more concerned about crime for others than for themselves. Results from our ordered logit models showed that concerns about crime for self and others were greater for respondents who felt less safe from crime and for respondents who used constrained behaviors. Personal concerns about crime were greater for women and concerns about crime for one’s spouse or significant other were greater for men. Contextual-level factors had little or no effect on concerns about crime. Regarding concerns about crime for one’s self and one’s spouse or significant other, there were also important differences by gender. For example, weapons behaviors mattered only for women and quality of life mattered only for men.

Keywords

fear of crime criminal victimization gender and crime/justice quantitative methods

Research on fear of crime has focused primarily on personal fear, or fear for oneself. Studies have shown that personal fear varies across demographic groups (e.g., women are more fearful than men; Ferraro, 1996; Reid & Konrad, 2004; Schafer, Huebner, & Bynum, 2006) and across contexts (e.g., fear is greater in more disorderly areas; Covington & Taylor, 1991; Franklin, Franklin, & Fearn, 2008; Scarborough, Like-Haislip, Novak, Lucas, & Alarid, 2010; Wyant, 2008). Less is known, however, about individuals’ fear for others (i.e., who they fear for, when they fear for them, and what the consequences are for themselves and others). Only a few studies have examined fear for others, and most have relied on data from a single city—for example, Dallas (Warr, 1992), Seattle (Drakulich & Rose, 2013), New York (Snedker, 2006), and Haifa, Israel (Mesch, 2000)—or state—for example, Texas (Warr & Ellison, 2000). Because these studies have focused almost exclusively on urban areas, less is known about fear for self and others in rural areas. Most studies have also focused on fear in familial relationships (e.g., Drakulich & Rose, 2013; Mesch, 2000; Warr & Ellison, 2000) and therefore the emphasis has been on individual-level predictors such as gender, marital status, and parental status. Although the context in which these relationships are embedded is likely also important, research examining how both individual- and contextual-level factors affect fear for others is sparse.

Studies consistently find that personal fear of crime affects individuals’ quality of life, for example by causing poor mental and physical health (e.g., anxiety, depression, reduced walking speed, and lung function; Norris & Kaniasty, 1992; Stafford, Chandola, & Marmot, 2007) and by constraining individuals’ behavior (Liska, Sanchirico, & Reed, 1988; Stafford et al., 2007). Although personal fear of crime matters more (or less) for specific subgroups, fear for others is more universal, such that even those who report little or no personal fear report fear for others. For example, Warr and Ellison’s (2000) survey of Texas households found that while 61% of married respondents reported personal fear, 84% reported fear for others (i.e., a spouse, son, or daughter). Fear for others is important, then, because it affects more people. Furthermore, fear for others may actually be more harmful (e.g., cause more distress) than fear for self, both because individuals likely fear for multiple others (e.g., intimate partners, children, family, and friends) and because they likely have less control over others’ safety than their own.

Efforts to explain fear of crime often distinguish between two types of factors: fear facilitators and fear inhibitors (Franklin et al., 2008; Schafer et al., 2006). Fear facilitators, factors that stimulate fear, include measures of vulnerability (e.g., gender, age, race, socioeconomic status, marital status, and prior victimization; Covington & Taylor, 1991; Killias & Clerici, 2000; Schafer et al., 2006) and disorder (e.g., incivilities such as groups of unsupervised teenagers, abandoned buildings and/or vehicles, and graffiti; Robinson, Lawton, Taylor, & Perkins, 2003; Wyant, 2008). Fear inhibitors, factors that protect against fear, include measures of social integration (e.g., how many neighbors you know by name, how often you talk with your neighbors, and feeling your neighborhood is more of a real home than just a place to live; Gibson, Zhao, Lovrich, & Gaffney, 2002). Regarding personal fear of crime, evidence suggests that perceptions of disorder matter most, followed by social integration and vulnerability (Franklin et al., 2008). It remains to be seen, however, whether and how these factors affect fear of crime for others.

Personal Fear of Crime

Fear of crime consists of three dimensions: (a) an affective dimension, defined as fear; (b) a cognitive dimension, defined as perceived risk; and (c) a behavioral dimension, defined as constrained behaviors (Hale, 1996). In contrast to fear, which focuses on worry about victimization, perceived risk focuses on assessments of one’s likelihood of becoming a victim (Ferraro & LaGrange, 1987). The distinction between fear and perceived risk is important because individuals may worry about victimization but believe that their risk of victimization is low, or vice versa. For this reason, scholars agree that fear of crime and perceived risk are conceptually distinct and should be measured separately. A related concern is that crime-specific fear measures are preferred over more general fear measures (Ferraro, 1995; Ferraro & LaGrange, 1987; LaGrange, Ferraro, & Supancic, 1992).

Constrained behaviors are the actions individuals take to protect themselves from potential victimization (Ferraro, 1995). The types of behaviors employed may include avoidance behaviors (e.g., avoiding certain people or places, limiting or changing one’s activities; Killias & Clerici, 2000; Liska et al., 1988; Stafford et al., 2007), protective behaviors (e.g., adding a security system or outside lighting, owning a watchdog; Ferguson & Mindel, 2007; Ferraro, 1996), and weapons behaviors (e.g., carrying a gun or other weapon; Rader & Haynes, 2014; Wilcox, May, & Roberts, 2006). Evidence suggests that the relationship between constrained behaviors and fear is reciprocal, such that fear increases constrained behaviors which increase fear (Liska et al., 1988). Although the focus of this study is on fear, it includes measures of both perceived risk and constrained behaviors. The following discussion highlights some of the key findings regarding individual and contextual effects on personal fear of crime.

Individual Characteristics

Individual characteristics associated with fear of crime include indicators of both physical vulnerability (e.g., gender and age) and social vulnerability (e.g., race, marital status, and prior victimization). According to the vulnerability hypothesis, fear of crime reflects individuals’ capacity for self-protection, such that individuals who feel less capable of defending themselves against a physical attack and/or whose lifestyles expose them to greater risk of victimization are more fearful (Franklin et al., 2008; Schafer et al., 2006). For example, studies have found that, compared to their respective counterparts, fear is greater among women and the elderly (Killias & Clerici, 2000; Warr, 1984) and among minorities (Covington & Taylor, 1991; Liska et al., 1988). The effect of gender is particularly salient, given that it is generally the strongest predictor of fear of crime (Killias & Clerici, 2000). Possible explanations for women’s greater fear of crime are that women feel more physically vulnerable to victimization (Killias & Clerici, 2000; Smith & Torstensson, 1997) and that their fear of sexual assault overshadows their fear of all other crimes (Ferraro, 1996; Fisher & Sloan, 2003; Warr, 1984; Wilcox, Jordan, & Pritchard, 2006).

With respect to age and fear, the findings are less consistent than those regarding gender. For example, studies have shown that older people are more fearful (Warr, 1984), that younger people are more fearful (Ferraro & LaGrange, 1992; Liu, Messner, Zhang, & Zhuo, 2009), and that a curvilinear relationship exists, such that both groups are more fearful than those in the middle of the age distribution (Ferraro, 1995; Warr & Ellison, 2000). One explanation for the latter finding is that both older and younger people are more physically vulnerable. Older people are more vulnerable because they feel less capable of defending themselves against a physical attack, in part because of issues related to their health (Beaulieu, Dube, Bergeron, & Cousineau, 2007; Cossman & Rader, 2011; Stiles, Halim, & Kaplan, 2003), and younger people are more vulnerable because of their smaller physical size (Goodey, 1994; May, 2001).

In contrast to gender and age, which reflect physical vulnerability, race reflects social vulnerability. Most studies have found that non-Whites are more fearful than Whites (Chiricos, Hogan, & Gertz, 1997; Haynie, 1998; Skogan & Maxfield, 1981), but some have found the opposite (Gainey, Alper, & Chappell, 2011; Rountree & Land, 1996). Minorities are considered more socially vulnerable because they live in communities where crime is more commonplace and therefore their risk of victimization is greater. Scarborough, Like-Haislip, Novak, Lucas, and Alarid (2010) found that although Blacks were more fearful than non-Blacks in their baseline model (which included only demographic characteristics), Blacks were actually less fearful than non-Blacks after controlling for neighborhood characteristics.

Other measures of vulnerability include marital status and prior victimization (Schafer et al., 2006). Regarding marital status, evidence suggests that married individuals are less personally fearful than are their non-married counterparts (Haynie, 1998). One reason for this finding may be that married individuals perceive themselves to be less vulnerable to victimization because they have another individual with whom to share their safety precautions (Rader, 2008). In terms of victimization, most studies have found that prior victimization is associated with increased fear of crime (Ferguson & Mindel, 2007; Kanan & Pruitt, 2002; May & Dunaway, 2000; Rountree & Land, 1996; Snedker, 2015). Most individuals are not victims of crime, however, and therefore their experiences with crime and victimization are more likely to be indirect (i.e., through friends, family, and/or the media; De Groof, 2008; Weitzer & Kubrin, 2004). For this reason, indirect victimization (i.e., hearing that a neighbor’s house had been burglarized) may matter more than direct victimization (Covington & Taylor, 1991).

Contextual Characteristics

Although fear of crime studies initially focused on the relationship between individual characteristics and fear, later studies have examined how contextual characteristics relate to fear levels. These contextual characteristics are sometimes measured at the individual level, by asking respondents about their perceptions of social disorder and social integration (Franklin et al., 2008; Gibson et al., 2002; Schafer et al., 2006). At other times, perceptions are aggregated to the neighborhood level (Covington & Taylor, 1991; Wyant, 2008) or to the street block level (Robinson et al., 2003).

Most studies have measured social disorder by asking respondents about perceived physical and social incivilities. Physical incivilities may include vandalism, graffiti, litter, and abandoned housing, while social incivilities may include groups of unsupervised teenagers, noisy neighbors, and public drinking (Covington & Taylor, 1991; Scarborough et al., 2010). Incivilities are threatening to residents because they represent a lack of community concern and an absence of informal social controls (Franklin et al., 2008). Consistent with this idea, most studies find that incivilities are associated with increased fear of crime (Covington & Taylor, 1991; Franklin et al., 2008; Liu et al., 2009; Markowitz, Bellair, Liska, & Liu, 2001; Scarborough et al., 2010; Wyant, 2008).

In contrast to social disorder, a fear facilitator, studies have also examined social integration, a fear inhibitor. Social integration refers to how connected individuals are to their community, such that individuals who are more socially integrated should report lower fear of crime levels (Franklin et al., 2008). Although most studies have found that social integration is associated with decreased fear of crime (Covington & Taylor, 1991; Franklin et al., 2008; Gibson et al., 2002), others have found that social integration and burglary-specific fear are positively related (Rountree & Land, 1996) or that social integration is relatively unimportant (Kanan & Pruitt, 2002). One reason for these disparate findings may be differences in how social integration has been measured—for example, with a single item focusing on neighbors’ willingness to call the police (Covington & Taylor, 1991), with multiple items including homeownership and neighborhood sentiment (Kanan & Pruitt, 2002), or with a scaled measure comprised of items such as talking with neighbors and feeling that one’s neighborhood is a real home (Franklin et al., 2008; Gibson et al., 2002) or recognizing strangers on one’s block and watching neighbors’ property when they are away (Rountree & Land, 1996).

Studies have also examined several objective (rather than perceptual) measures of community characteristics. Drawing on social disorganization theory, these measures include indicators of disadvantage (Markowitz et al., 2001; Scarborough et al., 2010), the racial composition of the community (Markowitz et al., 2001; Wyant, 2008), and community crime rates (Ferraro, 1996; Franklin et al., 2008; Liska et al., 1988; Scarborough et al., 2010; Wyant, 2008). Studies generally find that neighborhood disadvantage is associated with increased fear of crime (Scarborough et al., 2010). With respect to the racial composition of the community and community crime rates, however, the findings are mixed. For example, Covington and Taylor (1991) found that both individual- and aggregate-level measures of race were related to fear of crime. At the individual level, greater disparity between an individual’s race and the racial composition of his or her neighborhood related to greater fear of crime. At the aggregate level, respondents in predominately Black (i.e., greater than 90%) neighborhoods were more fearful. Others have found that it is perceived, rather than actual, racial composition that affects fear of crime, but only among Whites (i.e., Whites who perceive that they are in the racial minority in their community are more fearful; Chiricos et al., 1997). Most studies have found that crime rates have little or no effect on fear (Liska et al., 1988; Scarborough et al., 2010; Wyant, 2008). In some cases, however, crime rates matter only for certain crimes (i.e., burglary while at home and rape; Ferraro, 1996) or subgroups (i.e., Whites; Liska, Lawrence, & Sanchirico, 1982).

The relationship between community contexts and fear also differs by gender. Some evidence suggests that contextual characteristics matter more for women. For example, Smith and Torstensson (1997) claimed that women are more fearful because of their “ecological vulnerability” (i.e., because they perceive more risk in their communities). Consistent with this idea, Snedker (2015) found that social disorder and serious crime in the neighborhood mattered more for women’s fear of crime than for men’s. Other studies, however, suggest that community characteristics matter more for men. For example, Schafer, Huebner, and Bynum (2006) found that perceptions of disorder were associated with increased fear of crime, but men were more adversely affected by disorder than were women. One reason for these disparate findings may be differences in sample size and study location. Although Snedker’s (2015) research was based on a sample of 12,549 respondents from 12 U.S. cities, the studies by Schafer et al. (2006) and Smith and Torstensson (1997) were based on smaller samples (2,058 and 3,882, respectively) of respondents from a single city (a Midwestern U.S. city and Stockholm, Sweden, respectively). A second reason may be that these studies used different measures of contextual characteristics.

Fear of Crime for Others

In addition to personal fear, individuals report fear for others whose safety and well-being they value (Warr, 1992; Warr & Ellison, 2000). Compared to research on personal fear, however, research on fear for others, sometimes called altruistic fear (Warr, 1992) or vicarious fear (Snedker, 2006), is sparse. Furthermore, because most studies have focused on fear in the context of familial relationships (i.e., fear for a spouse or significant other and fear for children), the emphasis has been on individual-level predictors such as gender, marital status, and parental status. Little or no attention has been paid to contextual-level predictors of fear for others. In this article, we extend research on fear for others by examining both individual- and contextual-level predictors of fear for others.

There are two important differences between personal fear and fear for others. First, the relationship between personal fear and fear for others is gendered, such that women are more likely than men to fear for themselves, but men are more likely than women to fear for others (Snedker, 2006; Warr, 1992; Warr & Ellison, 2000). Furthermore, when women and men experience fear for others, they do so in different ways. For example, women who fear for others are more likely to fear for their children, while men who fear for others are more likely to fear for their spouse (Snedker, 2006; Warr, 1992; Warr & Ellison, 2000). This finding may be due, in part, to traditional family gender roles that associate caring for children with women and caring for spouses with men (Mesch, 2000; Rader, 2010; Snedker, 2006; Warr & Ellison, 2000). In other words, women are socialized to serve as caretakers and men are socialized to serve as protectors (Rader, 2010; Snedker, 2006). Although less pronounced today than in the past, these traditional gender roles persist and may explain gender differences in fear for others (Rader & Haynes, 2011). Second, although personal fear tends to increase with age (Warr, 1984), fear for others decreases with age (Warr, 1992; Warr & Ellison, 2000). The latter finding may be attributable to decreases in household size and therefore in the number of others one has to worry about (Warr, 1992).

Marriage also relates to fear for others; in general, married individuals are more likely than their non-married counterparts to fear for others (Warr, 1992). Furthermore, husbands are more likely than wives to report fear for their spouse. Warr and Ellison (2000) found that 47% of husbands and 33% of wives reported that they were very concerned about their spouse’s safety. Results from their logistic regression models showed that, among family households, men were significantly more likely than women to be very concerned about a spouse’s safety (i.e., the odds were 2.22 times higher for men than for women). Efforts to explain why husbands fear more than wives for their spouse often rely on gender socialization. For example, scholars have argued that husbands are socialized to protect their wives and that wives are socialized to rely on their husbands for protection (Rader, 2008, 2010; Snedker, 2006). For women, one benefit of marriage may be that they relinquish responsibility for their safety to their husbands (as part of the division of labor; Rader, 2008).

Parental status also relates to fear for others, such that parents are more likely than non-parents to fear for others (Warr, 1992). The relationship between parental status and fear for others also differs by gender. For example, Warr and Ellison (2000) found that parents feared more for their daughters than for their sons; the percentage of parents who reported being very concerned was higher for daughters (63%) than for sons (53%). Furthermore, female parents were slightly more likely than male parents to fear for their children. In the multivariate analyses, however, the difference in parents’ fear for their sons and daughters was not significant.

Evidence also suggests that women’s fear for children extends beyond their own children. For example, the women in Snedker’s (2006) study reported fearing both for their own children and for other neighborhood children. She concluded that women’s fear for others was more general than men’s because they were more likely to experience fear for a broad range of others (i.e., for both family and non-family members). The relationship between parental status and fear for children also varies by age. For example, younger women are more fearful than older women for their children (Mesch, 2000), in part because their children are younger and parents’ fear for children declines as their children grow older (Mesch, 2000; Warr & Ellison, 2000).

Although these findings provide important information about fear for others, only a few studies (e.g., Drakulich & Rose, 2013; Mesch, 2000; Snedker, 2006; Warr & Ellison, 2000) have examined the topic and several unanswered questions remain. For example, does fear for others extend beyond spouses and children? When do individuals fear for others? What are the consequences of that fear for themselves and others? Furthermore, most studies have focused exclusively on individual-level predictors of fear for others. Although contextual factors differentially affected men’s and women’s personal fear of crime, it is unclear whether the same is true for their fear for others.

The Current Study

The purpose of this study was to examine how individual- and contextual-level factors affected two proxies for fear of crime: respondents’ concerns about crime for themselves and others (i.e., their spouse or significant other, family, and friends). Specifically, we combined individual-level survey data from a national sample of U.S. adults with contextual-level official data from the U.S. Census, the American Community Survey (ACS), and the Uniform Crime Reports (UCR). At the individual level, we included measures of both fear facilitators (i.e., vulnerability) and fear inhibitors (i.e., social integration). At the contextual level, we included several indicators of social disorganization (i.e., characteristics of the social and economic contexts), which are also considered fear facilitators (see Franklin et al., 2008; Schafer et al., 2006).

We tested two separate hypotheses. First, we hypothesized that concerns about crime for others should vary according to the type of relationship (i.e., spouse or significant other, family, or friends). Specifically, respondents’ concerns about crime should be greatest for their spouse or significant other, followed by their family and friends. In other words, we expected that individuals would be more concerned about more intimate relationships. We also expected this relationship to vary by gender. That is, we expected women to be more concerned about crime for themselves and men to be more concerned about crime for others, particularly their spouse or significant other.

Second, we hypothesized that contextual characteristics should matter more for personal concerns about crime than for concerns about crime for others. Because individuals are less likely to have information about others’ contexts, we predicted that factors related to social integration and social disorganization should be more important for concerns about crime for self than for others. Given the gendered nature of fear for self and fear for one’s spouse or significant other, we were also interested in whether the effects of contextual characteristics differed by gender.

Method

Data

To examine individuals’ concerns about crime for themselves and for others (i.e., for their spouse or significant other, family, and friends), this study combined individual-level data from the National Survey of Attitudes toward Juvenile Justice, a 2009 telephone survey of U.S. adults, and contextual-level data from the 2010 U.S. Census, the 2008–2010 ACS, and the 2008–2010 UCR.

The telephone survey was divided into three sections and took approximately 15 minutes to complete. The first section asked a series of questions about the respondent’s community, including how long the respondent has lived in the community, whether the respondent knows the names of his or her neighbors, whether and how often the respondent has participated in unpaid volunteer activities, whether the respondent has attended community events, and the respondent’s satisfaction with his or her quality of life in the community. This section also included questions about how safe from crime and how concerned about crime the respondent felt in his or her home and neighborhood, as well as the types of behaviors the respondent employed because he or she was concerned about crime.

The second section asked a series of questions about the respondent’s criminal justice attitudes, including the importance of various purposes of punishment and the treatment of juvenile offenders. This section also included questions about the respondent’s attitudes about the future of juvenile justice policies and practices.

The third section asked about the respondent’s background, including questions about the respondent’s age, gender, race/ethnicity, marital status, number of children, education, socioeconomic status, religion, and previous victimization experiences.

Sampling procedures

Household telephone numbers from throughout the United States were selected using random-digit dialing (RDD) procedures¹ and one respondent was chosen at random within each household.² The sample of household telephone numbers included those not listed in public telephone directories. The response rate was 36%. We calculated the response rate by dividing the number of completed interviews (1,207) by the number of valid numbers (3,325). Valid numbers included completed interviews (1,207), refusals (518), incomplete callbacks (1,249), and communication problems (351). Of the remaining telephone numbers, 4,534 were not working numbers and 3,908 were no answer/busy signals. No answer/busy signals were considered ineligible because respondents’ eligibility to participate was unknown. Telephone numbers were dialed a maximum of 8 times. The final sample consisted of 1,207 U.S. adults.

Although our response rate is low by some standards, it is consistent with the response rates of other national RDD-based studies, which have declined significantly over time. For example, response rates for the Survey of Consumers have declined from about 70% in the 1980s to about 40% in recent years (Groves et al., 2009). According to the PEW Research Center (2012), the response rate for a typical telephone survey has declined from 36% in 1997 to 9% in 2012. Despite these declines, evidence suggests that nonresponse does not necessarily produce biased estimates. For example, Keeter, Miller, Kohut, Groves, and Presser (2000) compared two RDD surveys that used the same questions but different levels of effort: (a) a “standard” survey conducted over 5 days and based on a sample of adults who were home when the interviewer called and (b) a “rigorous” survey conducted over 8 weeks and based on a random selection of all adult household members. Response rates for the two surveys were 36% and 61%, respectively, but they produced similar results (e.g., respondents’ fear of neighborhood crime and familiarity or experience with criminal victimization did not differ significantly between the two types of surveys). Keeter, Kennedy, Dimcock, Best, and Craighill (2006) replicated this study a few years later and found that although response rates for the “standard” and “rigorous” surveys were lower (i.e., 25% and 50%, respectively), they again produced similar results.

Measures

Individual-level variables

Concerns about crime

We included one measure of respondents’ personal concerns about crime and three measures of respondents’ concerns about crime for others. To measure respondents’ personal concerns about crime, they were asked, “How worried are you about crime walking in your neighborhood?” Responses ranged from ‘1’ Not at all worried to ‘4’ Very worried. Respondents’ concerns about crime for others were based on the following questions: “How worried are you about crime for your spouse or significant other?” “How worried are you about crime for your family?” and “How worried are you about crime for your friends?” Responses ranged from ‘1’ Not at all worried to ‘4’ Very worried. These measures focus on general concerns about crime because those based on fear of specific crimes were not available in our data (Ferraro, 1995; Franklin et al., 2008; Rountree & Land, 1996; Warr, 2000).

Perceived risk

To measure respondents’ perceived risk of victimization, they were asked, “How safe from crime do you feel walking in your neighborhood?” Responses ranged from ‘1’ Not at all safe to ‘4’ Very safe.

Constrained behaviors

We included three types of constrained behaviors in this study: avoidance behaviors, protective behaviors, and weapons behaviors. For each type of constrained behavior, respondents were asked, “Have you ever…because you were concerned about crime in your community?” Avoidance behaviors was a dichotomous variable coded ‘1’ if respondents avoided specific areas during the day, avoided specific areas during the night, limited or changed their daily activities, or relied on others to walk or drive them to a location and ‘0’ otherwise. Protective behaviors was a dichotomous variable coded ‘1’ if respondents carried mace or pepper spray, took a self-defense class, added outside lighting or purchased deadbolt locks for their residence, owned a watchdog, or carried a cell phone for the purpose of safety and ‘0’ otherwise. Weapons behaviors was a dichotomous variable coded ‘1’ if respondents carried a knife, carried a firearm, or carried any type of weapon not mentioned and ‘0’ otherwise.

Respondent characteristics

To assess respondents’ vulnerability to crime, we included controls for respondents’ gender (coded ‘1’ for male and ‘0’ for female), age (in years), race (coded ‘1’ for White and ‘0’ for non-White), and marital status (coded ‘1’ for married and ‘0’ for non-married). We also controlled for whether respondents had children (coded ‘1’ for yes and ‘0’ for no) and were the victim of a crime in the past 12 months (coded ‘1’ for yes and ‘0’ for no). Our victimization measure was based on responses to the following two questions: “In the past 12 months, have you been the victim of a property crime such as vandalism, theft, or having your home broken into?” and “In the past 12 months, have you been the victim of a violent crime such as physical or sexual assault?”

We included four measures of social integration: respondents’ length of residence in the community (coded ‘1’ for less than 1 year, ‘2’ for 1–5 years, ‘3’ for 6–10 years, ‘4’ for 11–20 years, and ‘5’ for more than 20 years), whether respondents completed any unpaid volunteer activities or attended any community events in the past 12 months (coded ‘1’ for yes and ‘0’ for no), and respondents’ satisfaction with the quality of life in their community (coded ‘1’ for very dissatisfied, ‘2’ for dissatisfied, ‘3’ for satisfied, and ‘4’ very satisfied).

Contextual-level variables

We included three variables, taken from the 2010 U.S. Census, representing the social context of the county: the percentage of the population that was urban, the percentage of the population that was males between the ages of 15 and 24, and the percentage of the population that was Black. We also included one variable, based on a 3-year average from the ACS for the years 2008–2010, representing the economic context of the county: the percentage of the population that was living below the poverty level. Finally, we included an indicator of the county crime rate based on reported Part I offenses per 100,000 people, which came from the UCR. Specifically, we computed a 3-year average crime rate for the years 2008–2010. The crime rate represents a more objective measure of risk, as compared to the perceived risk variable.

Results

The results are in two parts. First, we present descriptive information about the sample of 1,207 U.S. adults (Table 1). We also compare men’s and women’s concerns about crime by calculating independent samples t-tests (Table 2). Second, we present results from the ordered logit models and describe how individual and contextual characteristics affect respondents’ concerns about crime for themselves, their spouse or significant other, and their family and friends (Table 3).³ We also estimate separate models predicting men’s and women’s personal concerns about crime and their concerns about crime for their spouse or significant other (Table 4). For both personal concerns about crime and concerns about crime for others, we estimated two models: the first model included only individual-level variables and the second model included both individual- and contextual-level variables. Because the effects of the individual-level variables were consistent across both models, we present only the full models here. To control for the fact that respondents are nested within different counties, we used the cluster procedure available in Stata. The cluster procedure corrects for within-county correlated errors (Helms & Jacobs, 2002).

Table 1.

Description of the Sample.

	Mean	SD	Range
Individual-level variables
Concerns about crime
For self	2.01	0.96	1.00	—	4.00
For spouse or significant other	2.05	0.99	1.00	—	4.00
For family	2.29	0.99	1.00	—	4.00
For friends	2.15	0.95	1.00	—	4.00
Perceived risk
Safe from crime	3.49	0.72	1.00	—	4.00
Constrained behaviors
Avoidance behaviors	0.49	0.50	0.00	—	1.00
Protective behaviors	0.81	0.39	0.00	—	1.00
Weapons behaviors	0.19	0.40	0.00	—	1.00
Respondent characteristics
Male	0.36	0.48	0.00	—	1.00
Age (in years)	55.32	16.41	18.00	—	96.00
White	0.84	0.37	0.00	—	1.00
Children	0.81	0.39	0.00	—	1.00
Married	0.58	0.49	0.00	—	1.00
Victimization	0.11	0.32	0.00	—	1.00
Length of residence	3.84	1.23	1.00	—	5.00
Volunteer	0.50	0.50	0.00	—	1.00
Community events	0.61	0.49	0.00	—	1.00
Quality of life	3.32	0.69	1.00	—	4.00
Contextual-level variables
Percentage urban	75.24	26.45	0.00	—	100.00
Percentage male, ages 15–24	7.11	1.54	4.10	—	20.30
Percentage black	12.09	12.88	0.10	—	73.50
Percentage living below poverty	10.23	4.21	2.20	—	30.50
Average crime rate (crimes per 100,000 people)	3,097.71	1,390.17	49.99	—	15,993.50

Note. N = 1,207.

Table 2.

Means on Concerns About Crime for Men and Women.

	Men	Women
	M	M	t-test
Concerns about Crime
For self	1.84	2.11	4.809***
For spouse or significant other	2.12	2.01	−1.637
For family	2.19	2.35	2.607**
For friends	2.09	2.19	1.759

*p < .05. **p < .01. ***p < .001.

Table 3.

Ordered Logistic Regression Models Predicting Concerns About Crime for Self and Others.

	Self		Spouse Or Significant Other		Family		Friends
	b	Robust SE	b	Robust SE	b	Robust SE	b	Robust SE
Individual-level variables
Perceived risk
Safe from crime	−1.24	0.13***	−0.68	0.13***	−0.48	0.11***	−0.45	0.12***
Constrained behaviors
Avoidance behaviors	0.60	0.13***	0.46	0.14***	0.72	0.13***	0.66	0.13***
Protective behaviors	0.34	0.19	0.64	0.20***	0.56	0.18**	0.79	0.17***
Weapons behaviors	0.48	0.15**	0.57	0.18**	0.68	0.15***	0.56	0.16***
Respondent characteristics
Male	−0.37	0.13**	0.44	0.14***	−0.12	0.13	−0.03	0.13
Age (in years)	−0.01	0.004	−0.01	0.005*	−0.01	0.004	−0.005	0.005
White	−0.11	0.17	−0.19	0.20	−0.55	0.19**	−0.34	0.19
Children	0.12	0.18	0.49	0.20*	0.53	0.17**	0.32	0.17
Married	−0.09	0.13	0.03	0.14	0.11	0.13	0.10	0.14
Victimization	0.04	0.20	−0.08	0.23	0.47	0.20*	0.35	0.19
Length of residence	0.03	0.05	−0.01	0.06	0.05	0.05	0.07	0.05
Volunteer	−0.09	0.13	−0.11	0.14	−0.11	0.13	−0.17	0.13
Community events	−0.13	0.14	−0.04	0.15	0.12	0.14	−0.02	0.14
Quality of life	−0.30	0.11**	−0.25	0.12*	−0.21	0.12	−0.21	0.11
Contextual-level variables
Percentage urban	0.003	0.003	−0.001	0.003	0.001	0.003	0.001	0.003
Percentage male, ages 15–24	−0.04	0.03	0.01	0.03	0.02	0.04	−0.01	0.03
Percentage black	0.002	0.01	0.004	0.01	−0.01	0.01	0.01	0.01
Percentage living below poverty	0.03	0.02	−0.01	0.02	−0.04	0.02*	−0.02	0.02
Crime rate (crimes per 100,000 people)	0.00003	0.00005	0.00001	0.0001	0.0001	0.0001	0.00004	0.00005
Cut 1	−5.60	0.72	−3.10	0.73	−2.61	0.68	−2.20	0.64
Cut 2	−3.99	0.70	−1.89	0.72	−1.39	0.67	−0.86	0.64
Cut 3	−1.50	0.67	0.31	0.71	1.00	0.67	1.69	0.63
N	1,039		869		1,008		1,029
Log pseudo likelihood	−1,104.90		−1,012.74		−1,201.52		−1,198.92
Wald χ²	339.90***		126.09***		163.58***		157.78***
Pseudo R ²	.15		.08		.08		.08

Note. SE = standard error.

*p < .05. **p < .01. ***p < .001.

Table 4.

Ordered Logistic Regression Models Predicting Concerns About Crime for Self and Spouse or Significant Other by Gender.

	Concerns About Crime for Self				Concerns About Crime for Spouse Or Significant Other
	Men		Women		Men		Women
	b	Robust SE	b	Robust SE	b	SE	b	Robust SE
Individual-level variables
Perceived risk
Safe from crime	−1.17	0.21***	−1.24	0.18***	−0.77	0.23***	−0.67	0.18***
Constrained behaviors
Avoidance behaviors	0.52	0.23*	0.65	0.16***	0.62	0.25*	0.39	0.17*
Protective behaviors	0.39	0.31	0.37	0.28	1.03	0.32***	0.43	0.29
Weapons behaviors	0.03	0.23	0.94	0.22***	0.39	0.25	0.70	0.28*
Respondent characteristics
Male	—	—	—	—	—	—	—	—
Age (in years)	−0.01	0.01	−0.004	0.01	−0.01	0.01	−0.01	0.01
White	0.02	0.32	−0.19	0.22	−0.23	0.35	−0.15	0.27
Children	0.34	0.29	0.02	0.22	0.67	0.33*	0.36	0.26
Married	−0.35	0.24	0.06	0.16	0.27	0.23	−0.08	0.20
Victimization	0.33	0.35	−0.12	0.25	0.16	0.40	−0.23	0.28
Length of residence	0.04	0.09	0.001	0.07	−0.002	0.10	−0.03	0.07
Volunteer	−0.17	0.22	−0.04	0.17	−0.11	0.22	−0.15	0.21
Community events	−0.15	0.22	−0.15	0.19	−0.12	0.23	0.04	0.21
Quality of life	−0.58	0.20**	−0.22	0.14	−0.43	0.20*	−0.19	0.17
Contextual-level variables
Percentage urban	0.01	0.01	−0.0004	0.004	0.01	0.01	−0.005	0.004
Percentage male, ages 15–24	−0.17	0.09	−0.02	0.04	0.04	0.11	0.01	0.03
Percentage black	−0.01	0.01	0.005	0.01	0.01	0.01	0.001	0.01
Percentage living below poverty	0.03	0.03	0.03	0.02	−0.01	0.03	−0.003	0.02
Crime rate (crimes per 100,000 people)	0.0001	0.0001	−0.00001	0.0001	0.00004	0.0001	−0.00003	0.0001
Cut 1	−6.08	1.37	−5.39	0.86	−3.09	1.44	−3.68	0.91
Cut 2	−4.67	1.35	−3.64	0.84	−1.81	1.42	−2.48	0.89
Cut 3	−2.07	1.31	−1.15	0.80	0.88	1.41	−0.50	0.87
N	375		664		323		546
Log pseudo likelihood	−374.81		−716.68		−359.39		−641.61
Wald χ²	125.90***		218.52***		77.62***		62.21***
Pseudo R ²	.15		.16		.12		.07

*p < .05. **p < .01. ***p < .001.

Descriptive Analyses

Table 1 indicates that the majority of the 1,207 respondents in our sample were female (64%), White (84%), had children (81%), and were married (58%). The average respondent was 55 years of age. Compared to the 2010 U.S. Census, our sample has more women (51% vs. 64%, respectively), more Whites (72% vs. 84%, respectively), and is older (a median age of 37 years vs. 55 years, respectively; U.S. Census Bureau, 2010). Most respondents were not previously victimized (89%).⁴ Regarding their concerns about crime, respondents reported being most worried about crime for their family (M = 2.29), followed by their friends (M = 2.15), their spouse or significant other (M = 2.05), and themselves (M = 2.01). Respondents reported feeling safe (M = 3.49), indicating that their perceived risk of victimization was low. Regarding constrained behaviors, respondents were most likely to engage in protective behaviors (81%), followed by avoidance behaviors (49%) and weapons behaviors (19%).

With respect to the contextual-level variables, Table 1 shows that the average percentage of the population that was urban was 75%, the average percentage of the population that was males between the ages of 15 and 24 was 7%, the average percentage of the population that was Black was 12%, and the average percentage of the population living below the poverty level was 10%. The average crime rate was 3,098 crimes per 100,000 people.

Given the significant relationship between gender and fear of crime in previous studies, we also estimated t-tests comparing men’s and women’s concerns about crime for themselves and others. The results, presented in Table 2, show that concerns about crime for one’s self (t = 4.809, p < .001) and concerns about crime for one’s family (t = 2.607, p < .01) were significantly higher for women than for men. Concerns about crime for friends were higher among women and concerns about crime for one’s spouse or significant other were higher among men, but these differences were not significant.

Regression Analyses

Concerns about crime for self

Table 3 presents the results of the ordered logit models predicting personal concerns about crime. Personal concerns about crime were greater for respondents who felt less safe from crime (i.e., greater perceived risk), for respondents who used avoidance and weapons behaviors, for women, and for respondents who were less satisfied with the quality of life in their community. Contextual-level variables had no effect on personal concerns about crime.

Concerns about crime for others

Table 3 also presents the results of the analyses predicting concerns about crime for one’s spouse or significant other, family, and friends. Consistent with personal concerns about crime, concerns about crime for others were greater for respondents who felt less safe from crime and for respondents who used constrained behaviors. Unlike personal concerns about crime, however, all three types of behaviors—avoidance, protective, and weapons—mattered for concerns about crime for others. Concerns about crime for one’s spouse or significant other were also greater for men, for younger respondents, for respondents with children, and for respondents who were less satisfied with the quality of life in their community. Contextual-level variables had no effect on concerns about crime for one’s spouse or significant other.

Concerns about crime for one’s family were greater for non-White respondents, for respondents with children, for respondents who reported having been victimized in the past 12 months, and for respondents from counties with a lower percentage of the population living below the poverty level. Only perceived risk and constrained behaviors predicted concerns about crime for one’s friends.

Gender differences in concerns about crime

As stated previously, the relationship between fear for self and others, particularly one’s spouse or significant other, is gendered. Given this finding, we also estimated personal concerns about crime and concerns about crime for one’s spouse or significant other separately by gender (see Table 4). The purpose of these models was to examine whether and how individual- and contextual-level variables differentially affected men and women.

Among both men and women, concerns about crime (both for one’s self and for one’s spouse or significant other) were greater for respondents who felt less safe from crime and for respondents who used avoidance behaviors. Furthermore, weapons behaviors mattered for women (i.e., concerns about crime were greater for women who used weapons behaviors) and quality of life mattered for men (i.e., concerns about crime were greater for men who were less satisfied with the quality of life in their community). Among men, concerns about crime for one’s spouse or significant other were also greater for respondents who used protective behaviors and for respondents with children. Consistent with the overall models, contextual-level variables had no effect on concerns about crime among men and among women.

Summary

In sum, the results showed that respondents worried most about crime for their family, followed by their friends, their spouse or significant other, and themselves. Concerns about crime for one’s self and one’s family were significantly higher for women than for men. In the multivariate analyses, both personal concerns about crime and concerns about crime for others were greater for respondents who felt less safe from crime and for respondents who used constrained behaviors. Consistent with previous studies, personal concerns about crime were greater among women and concerns about crime for one’s spouse or significant other were greater among men. The contextual-level variables had little or no effect on concerns about crime. Only the percentage of the population living below the poverty level mattered, and this variable mattered only for concerns about crime for one’s family. There were important differences by gender, such that concerns about crime for one’s self and one’s spouse or significant other were greater for women who used weapons behaviors and for men who were less satisfied with the quality of life in their community.

Discussion

Most fear of crime research has focused primarily on personal concerns about crime. As a result, less is known about the nature and extent of fear for others. To address this shortcoming, this study used data from a national sample of U.S. adults to examine how individual- and contextual-level factors affected respondents’ personal concerns about crime and their concerns about crime for others (i.e., their spouse or significant other, family, and friends). Contrary to our first hypothesis, respondents worried most about crime for their family, followed by their friends, their spouse or significant other, and themselves. As expected, however, there were significant differences by gender; women were more likely than men to report concerns about crime for their family and for themselves. Family may have been most important because respondents were considering their immediate family (e.g., children) rather than their extended family (e.g., parents, siblings, and grandparents). More generally, however, the results are consistent with Warr and Ellison’s (2000) finding that individuals fear more for others than for themselves. Furthermore, they highlight the importance of examining different types of others. Although most studies have focused on fear in familial relationships (e.g., Drakulich & Rose, 2013; Mesch, 2000; Warr & Ellison, 2000), the finding that respondents feared more for their friends than for their spouse or significant other indicates that non-familial relationships are also important. To better investigate these familial and non-familial relationships, future research should include fear for specific types of others, especially family (e.g., intimate partners, children, parents, siblings, and grandparents) and friends (e.g., close friends and acquaintances).

An analysis of the effects of contextual characteristics on both personal concerns about crime and concerns about crime for others was a key objective of this study. Results from the multivariate analyses showed that contextual characteristics had little or no effect on concerns about crime. Only respondents' satisfaction with the quality of life in their community (a measure of social integration) and the percentage of the population living below the poverty level affected concerns about crime. Specifically, respondents who were less satisfied with their quality of life were more concerned about crime for themselves and for their spouse or significant other. Furthermore, respondents in counties with a lower percentage of the population living below the poverty level reported greater concerns about crime for family. One reason for this finding may be that respondents in these counties have more to lose (i.e., because they are less disadvantaged).

Results also showed that both personal concerns about crime and concerns about crime for others were greater for respondents who felt less safe from crime and for respondents who used constrained behaviors. As expected, men were less likely to report personal concerns about crime but more likely to report concerns about crime for their spouse or significant other. Finally, respondents with children were more likely to report concerns about crime for their spouse or significant other and for their family.

Given the gendered nature of fear for self and fear for one’s spouse or significant other, we also estimated these models separately for men and women. Results showed that quality of life mattered only for men. One explanation for this finding may be that men who are less satisfied with the quality of their lives are more likely to worry about a variety of issues, including crime and victimization. Because women are more fearful than men overall, their quality of life may be less likely to affect their fear levels. Although constrained behaviors mattered for men and women, the types of behaviors they employed differed by gender. For example, weapons behaviors mattered only for women. Because gun ownership, both for general purposes and for protection, is more common among men (Cao, Cullen, & Link, 1997), women who use weapons behaviors may be especially fearful of crime. The results also showed that protective behaviors predicted concerns about crime for one’s spouse or significant other but only among men. Protective behaviors may have mattered more for men because they are socialized to serve as protectors (Rader, 2010; Snedker, 2006).

Limitations

Although we believe this research contributes to the literature on concerns about crime, there are a few limitations worth noting. First, our measures of personal concerns about crime and concerns about crime for others were only proxies for fear. We relied on a general measure of concerns about crime, but evidence suggests that measures based on crime-specific fears (e.g., fear of sexual assault and burglary) are more appropriate (Ferraro, 1995; Franklin et al., 2008; Rountree & Land, 1996; Warr, 2000). A related concern is that our measures of concerns about crime asked respondents about how “worried” they were rather than how “fearful” they were. Although similar emotions, some respondents (e.g., men) may be more likely to report being “worried” than being “afraid.” The secondary nature of our data therefore limited our operationalization of concerns about crime (i.e., fear).

Second, our measures of concerns about crime for others did not distinguish between specific types of others. For example, our measure of concerns about crime for family did not distinguish between concerns for children and concerns for other family members (e.g., parents, siblings, and grandparents). It is also likely that concerns about crime for others depend on others’ personal characteristics. For example, future research should examine more nuanced measures that assess concerns about crime for male versus female children, for adolescent versus adult children, for parents versus grandparents, and for close friends versus acquaintances.

Third, women and older individuals, who are generally more fearful of crime (Killias & Clerici, 2000; Warr, 1984), were overrepresented in our data. Although the overrepresentation of women is common to many telephone surveys, older individuals may have been overrepresented, in part, because we sampled household telephone numbers, thereby excluding respondents who used only a cell phone. The rapid growth of cell phone only households is of particular importance because these respondents are more likely than those from households with landline telephones to be young and poor (Blumberg & Luke, 2007). By sampling household telephone numbers, we also excluded respondents who are homeless and who could not afford a telephone. Taken together, these limitations suggest that the results of this study may underestimate concerns about crime among respondents who are male, young, and poor.

A fourth limitation is our use of the UCR to measure county crime rates. The UCR has a number of well-documented limitations (see Loftin & McDowall, 2010), including the fact that it is restricted to only those crimes that are reported to the police, that it counts crimes according to the hierarchy rule (meaning that only the most serious offense in a criminal situation is recorded), and that its definition of rape excludes men.⁵ Nevertheless, the UCR is the only source of crime data available in many areas and therefore serves as a valuable, albeit imperfect, tool for estimating the effects of crime rates on fear.

Conclusion

Using data from a national sample of U.S. adults, this article examined how individual- and contextual-level factors affected both personal concerns about crime and concerns about crime for one’s spouse or significant other, family, and friends. Consistent with others (e.g., Mesch, 2000; Warr & Ellison, 2000), we found that individuals were more concerned about crime for others than for themselves. Furthermore, we found that contextual-level variables had little or no effect on concerns about crime. One reason for this finding may be that concerns about crime are more about who others are than about where they reside. In other words, when individuals fear for others, they fear for them across all contexts. A second reason for this finding may be that we relied on county-level measures, rather than neighborhood-level measures. It may be that individuals’ immediate environment matters more for their concerns about crime. For example, neighborhood characteristics are likely more important than county characteristics when parents are reporting concerns about crime for their children. For this reason, future research should examine a more diverse range of neighborhood and/or street block-level variables on concerns about crime for one’s self and others.

The results also showed that there were important differences by gender. Specifically, personal concerns about crime were greater for women and concerns about crime for one’s spouse or significant other were greater for men. Furthermore, there were differences in the types of constrained behaviors they employed, such that weapons behaviors mattered only for women and protective behaviors mattered only for men (but only for one’s spouse or significant other). These findings highlight the importance of educating people about their own and others’ risk of victimization. Providing accurate information about individuals’ objective risk is important, in part, because of the relationship between constrained behaviors and concerns about crime. This information may be particularly important for women, given that concerns about crime were greater for those who reported using weapons behaviors. Taken together, the results highlight the need for additional research examining the effects of individual and contextual characteristics on both personal concerns about crime and concerns about crime for others.

Footnotes

Acknowledgment

We thank Lindsay Jones for her assistance designing the telephone survey.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research was supported by funds from the Social Science Research Center at Mississippi State University.

Notes

References

Beaulieu

Dube

Bergeron

Cousineau

M.-M.

(2007). Are elderly men worried about crime? Journal of Aging Studies, 21, 336–346.

Blumberg

S. J.

Luke

J. V.

(2007). Coverage bias in traditional telephone surveys of low-income and young adults. Public Opinion Quarterly, 71, 734–749.

Cao

Cullen

F. T.

Link

B. G.

(1997). The social determinants of gun ownership: Self-protection in an urban environment. Criminology, 35, 629–657.

Chiricos

Hogan

Gertz

(1997). Racial composition of neighborhood and fear of crime. Criminology, 35, 107–131.

Cossman

J. S.

Rader

N. E.

(2011). Fear of crime and personal vulnerability: Examining self-rated health. Sociological Spectrum, 31, 141–162.

Covington

Taylor

R. B.

(1991). Fear of crime in urban residential neighborhoods: Implications of between- and within-neighborhood sources for current models. The Sociological Quarterly, 32, 231–249.

De Groof

(2008). And my mama said…The (relative) parental influence on fear of crime among adolescent girls and boys. Youth & Society, 39, 267–293.

Drakulich

K. M.

Rose

(2013). Being male or living with a female: Fear for partners by sex and sexual orientation. Journal of Interpersonal Violence, 28, 1765–1797.

Federal Bureau of Investigation. (2013). Reporting rape in 2013: Summary Reporting System (SRS) user manual and technical specification (C. J. I. S. C. Division, Trans.). Washington, DC: U.S. Department of Justice.

10.

Ferguson

K. M.

Mindel

C. H.

(2007). Modeling fear of crime in Dallas neighborhoods: A test of social capital theory. Crime & Delinquency, 53, 322–349.

11.

Ferraro

K. F.

(1995). Fear of crime: Interpreting victimization risk. New York, NY: SUNY Press.

12.

Ferraro

K. F.

(1996). Women's fear of victimization: Shadow of sexual assault? Social Forces, 75, 667–690.

13.

Ferraro

K. F.

LaGrange

(1987). The measurement of fear of crime. Sociological Inquiry, 57, 70–101.

14.

Ferraro

K. F.

LaGrange

R. L.

(1992). Are older people most afraid of crime? Reconsidering age differences in fear of victimization. Journal of Gerontology: Social Sciences, 47, S233–S244.

15.

Fisher

B. S.

Sloan

III, J. J.

(2003). Unraveling the fear of sexual victimization among college women: Is the “shadow of sexual assault” hypothesis supported? Justice Quarterly, 20, 633–659.

16.

Franklin

T. W.

Franklin

C. A.

Fearn

N. E.

(2008). A multilevel analysis of the vulnerability, disorder, and social integration models of fear of crime. Social Justice Research, 21, 204–227.

17.

Gainey

Alper

Chappell

A. T.

(2011). Fear of crime revisited: Examining the direct and indirect effects of disorder, risk perception, and social capital. American Journal of Criminal Justice, 36, 120–137.

18.

Gibson

C. L.

Zhao

Lovrich

N. P.

Gaffney

M. J.

(2002). Social integration, individual perceptions of collective efficacy, and fear of crime in three cities. Justice Quarterly, 19, 537–564.

19.

Goodey

(1994). Fear of crime: What can children tell us? International Review of Victimology, 3, 195–210.

20.

Groves

R. M.

Fowler

F. J.

Jr Couper

M. P.

Lepkowski

J. M.

Singer

Tourangeau

(2009). Survey methodology. Hoboken, NJ: John Wiley & Sons.

21.

Hale

(1996). Fear of crime: A review of the literature. International Review of Victimology, 4, 79–150.

22.

Haynie

D. L.

(1998). The gender gap in fear of crime, 1973-1994: A methodological approach. Criminal Justice Review, 23, 29–50.

23.

Helms

Jacobs

(2002). The political context of sentencing: An analysis of community and individual determinants. Social Forces, 81, 577–604.

24.

Kanan

J. W.

Pruitt

M. V.

(2002). Modeling fear of crime and perceived victimization risk: The (in)significance of neighborhood integration. Sociological Inquiry, 72, 527–548.

25.

Keeter

Kennedy

Dimcock

Best

Craighill

(2006). Gauging the impact of growing nonresponse on estimates from a national RDD telephone survey. Public Opinion Quarterly, 70, 759–779.

26.

Keeter

Miller

Kohut

Groves

R. M.

Presser

(2000). Consequences of reducing nonresponse in a national telephone survey. Public Opinion Quarterly, 64, 125–148.

27.

Killias

Clerici

(2000). Different measures of vulnerability in their relation to different dimensions of fear of crime. British Journal of Criminology, 40, 437–450.

28.

LaGrange

R. L.

Ferraro

K. F.

Supancic

(1992). Perceived risk and fear of crime: Role of social and physical incivilities. Journal of Research in Crime and Delinquency, 29, 311–334.

29.

Liska

A. E.

Lawrence

J. J.

Sanchirico

(1982). Fear of crime as a social fact. Social Forces, 60, 760–770.

30.

Liska

A. E.

Sanchirico

Reed

M. D.

(1988). Fear of crime and constrained behavior specifying and estimating a reciprocal effects model. Social Forces, 66, 827–837.

31.

Liu

Messner

S. F.

Zhang

Zhuo

(2009). Socio-demographic correlates of fear of crime and the social context of contemporary urban china. American Journal of Community Psychology, 44, 93–108.

32.

Loftin

McDowall

(2010). The use of official records to measure crime and delinquency. Journal of Quantitative Criminology, 26, 527–532.

33.

Markowitz

F. E.

Bellair

P. E.

Liska

A. E.

Liu

(2001). Exending social disorganization theory: Modeling the relationships between cohesion, disorder, and fear. Criminology, 39, 293–319.

34.

May

D. C.

(2001). The effect of fear of sexual victimization on adolescent fear of crime. Sociological Spectrum, 21, 141–174.

35.

May

D. C.

Dunaway

R. G.

(2000). Predictors of fear of criminal victimization at school among adolescents. Sociological Spectrum, 20, 149–168.

36.

Mesch

(2000). Women's fear of crime: The role of fear for the well-being of significant others. Violence & Victims, 15, 323–336.

37.

Norris

F. H.

Kaniasty

(1992). A longitudinal study of the effects of various crime prevention strategies on criminal victimization, fear of crime, and psychological distress. American Journal of Community Psychology, 20, 625–648.

38.

Pew Research Center. (2012). Assessing the representativeness of public opinion surveys. Washington, DC: Pew Research Center.

39.

Rader

N. E.

(2008). Gendered fear strategies: Intersections of doing gender and fear management strategies in married and divorced women's lives. Sociological Focus, 41, 34–52.

40.

Rader

N. E.

(2010). Until death do us part? Husband perceptions and responses to fear of crime. Deviant Behavior, 31, 33–59.

41.

Rader

N. E.

Haynes

S. H.

(2011). Gendered fear of crime socialization: An extension of Akers's social learning theory. Feminist Criminology, 6, 291–307.

42.

Rader

N. E.

Haynes

S. H.

(2014). Avoidance, protective, and weapons behaviors: An examination of constrained behaviors and their impact on concerns about crime. Journal of Crime and Justice, 37, 197–213.

43.

Reid

L. W.

Konrad

(2004). The gender gap in fear: Assessing the interactive effects of gender and perceived risk on fear of crime. Sociological Spectrum, 24, 399–425.

44.

Robinson

J. B.

Lawton

B. A.

Taylor

R. B.

Perkins

D. D.

(2003). Multilevel longitudinal impacts of incivilities: fear of crime, expected safety, and block satisfaction. Journal of Quantitative Criminology, 19, 237–274.

45.

Rountree

P. W.

Land

K. C.

(1996). Perceived risk versus fear of crime: Empirical evidence of conceptually distinct reactions in survey data. Social Forces, 74, 1353–1376.

46.

Scarborough

B. K.

Like-Haislip

T. Z.

Novak

K. J.

Lucas

W. L.

Alarid

L. F.

(2010). Assessing the relationship between individual characteristics, neighborhood context, and fear of crime. Journal of Criminal Justice, 38, 819–826.

47.

Schafer

J. A.

Huebner

B. M.

Bynum

T. S.

(2006). Fear of crime and criminal victimization: Gender-based contrasts. Journal of Criminal Justice, 34, 285–301.

48.

Skogan

W. G.

Maxfield

M. G.

(1981). Coping with crime: Individual and neighborhood reactions (Vol. 124). Beverly Hills: CA: Sage.

49.

Smith

W. R.

Torstensson

(1997). Gender differences in risk perception and neutralizing fear of crime. British Journal of Criminology, 37, 608–634.

50.

Snedker

K. A.

(2006). Altruistic and vicarious fear of crime: Fear for others and gendered social roles. Sociological Forum, 21, 163–195.

51.

Snedker

K. A.

(2015). Neighborhood conditions and fear of crime: A reconsideration of sex differences. Crime & Delinquency, 61, 45–70.

52.

Stafford

Chandola

Marmot

(2007). Association between fear of crime and mental health and physical functioning. American Journal of Public Health, 97, 2076–2081.

53.

Stiles

B. L.

Halim

Kaplan

H. B.

(2003). Fear of crime among individuals with physical limitations. Criminal Justice Review, 28, 232–253.

54.

U.S. Census Bureau. (2010). Profile of general population and housing characteristics: 2010. Retrieved January 21, 2014, from http://factfinder2.census.gov/rest/dnldController/deliver?_ts=408373145282

55.

Warr

(1984). Fear of victimization: Why are women and the elderly more afraid? Social Science Quarterly, 65, 681–702.

56.

Warr

(1992). Altruistic fear of victimization in households. Social Science Quarterly, 73, 723–736.

57.

Warr

(2000). Fear of crime in the United States: Avenues for research and policy. Criminal Justice, 4, 451–489.

58.

Warr

Ellison

C. G.

(2000). Rethinking social reactions to crime: Personal and altruistic fear in family households. American Journal of Sociology, 106, 551–578.

59.

Weitzer

Kubrin

C. E.

(2004). Breaking news: How local TV news and real-world conditions affect fear of crime. Justice Quarterly, 21, 497–520.

60.

Wilcox

Jordan

C. E.

Pritchard

A. J.

(2006). Fear of acquaintance versus stranger rape as a master status: Towards refinement of the shadow of sexual assault. Violence & Victims, 21, 355–370.

61.

Wilcox

May

D. C.

Roberts

S. D.

(2006). Student weapon possession and the “fear and victimization hypothesis”: Unraveling the temporal order. Justice Quarterly, 23, 502–529.

62.

Williams

(2006). Generalized ordered logit/partial proportional odds models for ordinal dependent variables. The Stata Journal, 6, 58–62.

63.

Wyant

(2008). Multilevel impacts of perceived incivilities and perceptions of crime risk on fear of crime. Journal of Research in Crime and Delinquency, 45, 39–64.